Introduction

Hypervalent iodine(III) reagents have gained attention as strong oxidants with a low toxicity and due to the ability to mimic reactivity usually associated with transition metals . Iodine(III) compounds have been used for the formation of different bond types, such as C–C , C–O , C–N , C–S , C–CN , C–F , C–I , C–NO2 and, in the context of this work, C–X (X = Cl, Br) . So far, different protocols for the halogenation of arenes using iodine(III) reagents have been described, mainly using (diacetoxyiodo)benzene (PIDA)/TMSCl, PIDA/TMSBr , and [bis(trifluoroacetoxy)iodo]benzene (PIFA)/TMSBr . We have recently developed a new protocol for the oxidative chlorination and bromination of naphthols using the PIFA/AlCl3 and PIDA/AlBr3 systems. These unprecedented protocols combined iodine(III) reagents and aluminum salts to achieve chlorination and bromination of electron-rich arenes under mild and experimentally straightforward conditions (Scheme 1).

Scheme 1: Representative protocols for the oxidative aromatic chlorination and bromination with iodine(III) reagents.

The synthesis of aryl halides is of great academic and industrial importance. Recently, our research group has developed a new procedure for the ortho-selective chlorination of phenols under mild conditions in a short reaction time . The chlorinating species was generated in situ simply by mixing PIFA with a Lewis acid, in this case AlCl3. The importance of this protocol arises from the oxidation of an AlCl3-based chlorine atom, which is an available and cheap reagent. Then it is used as an electrophile source in the chlorination process with an umpolung reactivity. In contrast to the suggested traceroute where the chlorine or bromine atom is attached to the hypervalent iodine center of the plausible reagent PhIX2 (X = Cl, Br), our new protocol opens up a broad path for the reaction through different halogenating species. For a deeper understanding of these reactions, we explored different pathways of the reaction mechanisms for the ortho-halogenation using 2-naphthol as a model substrate (Scheme 2). In such a way, we found a reaction pathway that was energetically favored.

Scheme 2: Chlorination of 2-naphthol using the PIFA/AlCl3, 1:2 system.

Based on our successful procedure for chlorination, we also developed an efficient protocol for the electrophilic bromination of arenes, mainly phenols . Accordingly, the bromination reaction was initially explored by mixing PIFA and AlBr3, which gave an acceptable yield (84%). However, other iodine(III) reagents were tested as oxidants during the optimization process. Thus, when mixing PIDA with aluminum bromide, the reaction occurred with an unexpectedly higher yield (93%) than with PIFA. Therefore, the bromination reaction proceeded in the presence of PIDA/AlBr3 as a brominating system using MeCN as solvent (Scheme 3).

Scheme 3: Bromination of 2-naphthol using the PIDA/AlBr3, 1:2 system.

In light of the relevance of this newly discovered reactivity and the scarce mechanistic and theoretical studies available , we computationally explored all of the different plausible pathways to elucidate the most feasible route that allowed the reported halogenation under these new reaction conditions. In this work, we systematically investigated the influence of PIDA and PIFA in the chlorination and bromination reactions. Interestingly, we found an excellent agreement between the theoretical predictions and the experimental results.

Results and Discussion Computational details

The equilibrium geometry of reagents and products, the stationary points, and transition-state structures were optimized by density functional theory (DFT) calculations employing the software Gaussian 16 . Although the B3LYP functional could be suitable for these calculations, e.g., for tracing reaction pathways, nitrations, halogenations, or FC acylations in solution, we found the ω-B97XD functional appropriate for this study because it considered dispersion interactions through a range separation (22% for short range and 100% Hartree–Fock for long range), which properly describes thermochemistry and noncovalent interactions. When searching for the critical points along the potential surface energy of the possible chlorination and bromination pathways studied in this work, Br and I atoms were treated with the revised version of the LANL2DZ basis set and effective core potential, referred to as LANL08(d), providing d-type polarization functions. Meanwhile, the 6-31G(d) basis set was used for the other atoms (i.e., H, C, O, F, Al, etc.).

Geometry optimizations were carried out without any symmetry constraints, and the stationary points were characterized by analytical frequency calculations, i.e., energy minima (reactants, intermediates, and products) must exhibit only positive harmonic frequencies, whereas each energy maximum (transition state) exhibited only one negative frequency. From these last calculations, zero-point energy, thermal, and entropy corrections were obtained, which were added to the electronic energy to express the calculated values as Gibbs free energy at 298 K and 1 atm.

All our calculations were performed in the gas phase. Then, the solvent effects were included according to the polarizable continuum model via the solvent model density (SMD) option considering Truhlar’s model and MeCN as the solvent. Single-point calculations were improved using a mixed basis set of triple-ζ quality with a polarization function, 6-311G(d,p) for all atoms except for Br and I, which were treated with the LANL08d relativistic pseudopotential , i.e., the composite level of theory used is the following: (SMD: MeCN) ω-B97XD/(6-311G(d,p),LANL08d)//ω-B97XD/6-31G(d), LANL08d.

Chlorination mechanism

The reaction mechanism for the chlorination of 2-naphthol using one equivalent of PIFA and two equivalents of aluminum chloride is outlined in Scheme 4.

Scheme 4: Reaction mechanism for the chlorination of 2-naphthol using the PIFA/AlCl3, 1:2 system.

The chlorination mechanism starts when PIFA coordinates the first equivalent of aluminum chloride to give the corresponding adduct PIFA–AlCl3. Next, a chlorine atom is transferred to the iodine(III) center to yield I-1–Cl via TS1–Cl with the release of the complex TFAO–AlCl2. Then, the second equivalent of aluminum chloride coordinates the TFAO ligand, giving rise to the chlorinating species I-2–Cl in equilibrium with I-3–Cl. At this point, 2-naphthol reacts, leading to the formation of the ion pair I-4–Cl via chlorine atom transfer, which then yields the adduct I-5–Cl trough transition state TS2–Cl. Then, the release of the second equivalent of the TFAO–AlCl2 complex yields the final product 1-chloro-2-naphthol (P–Cl).

The calculated mechanism for the chlorination reaction starts with coordination of a PIFA oxygen atom to aluminum chloride. This generates a highly exergonic PIFA–AlCl3 adduct. In Figure 1, the Gibbs free energy of this adduct is set as 0 kcal/mol for more clarity. Herein, one chlorine atom is transferred from aluminum to the hypervalent iodine(III) center through six-membered-ring transition state TS1–Cl (ΔG‡ = 9.7 kcal/mol, selected bond lengths 2.76, 1.22, 1.27, 1.78, 2.60, and 2.86 Å for I–O, O–C, C–O, O–Al, Al–Cl, and Cl–I, respectively). Then, the tetracoordinate TFAO–AlCl2 salt is released, giving rise to intermediate I-1–Cl (ΔG = −25.2 kcal/mol), which contains the key Cl–I(III) bond, in a formal TFAO/Cl ligand exchange. The Cl–I bond length is 2.46 Å, with the halogen atom sharing the hypervalent iodine bond in the equatorial position. Next, the second equivalent of aluminum chloride coordinates to the TFAO ligand, forming active chlorinating species I-2–Cl (ΔG = −18.3 kcal/mol). This energetically favored step is in equilibrium with the ion pair I-3–Cl (ΔG = −0.5 kcal/mol). It is worth mentioning that the slight difference in energy between both states indicates the importance of the spontaneous interconversion of both species, which is observed only in the presence of two equivalents of the Lewis acid.

Figure 1: Energy profile for the chlorination of 2-naphthol in the presence of PIFA and AlCl3.

After the addition of 2-naphthol, the chlorine atom is introduced barrier-free into the phenolic ring, producing the nonaromatic intermediate I-4–Cl (ΔG = −23.1 kcal/mol). Next, aromatization assisted by TFAO–AlCl2 via TS2–Cl (ΔG‡ = 11.1 kcal/mol) and hydrogen transfer from the nonaromatic intermediate to TFAO–AlCl2 are observed. In TS2–Cl, the energy barrier must be overcome to give rise to the 1-chloro-2-naphthol adduct with TFA–OH–AlCl2, I-5–Cl (ΔG = −2.6 kcal/mol), which spontaneously yields the final 1-chloro-2-naphthol (P–Cl) with concomitant release of TFAO–AlCl2 in a highly exothermic process (ΔG = −44.2 kcal/mol, Figure 1).

Other relevant routes for this chlorination process, which involve a different stoichiometry or the formation of PhICl2 as chlorinating species, were also investigated and ruled out. Thus, for the chlorination of 2-naphthol with the PIFA/AlCl3, 1:1 system, we found that in general, once the intermediate I-1–Cl is formed, the following coordination of 2-naphthol with the TFAO ligand via TS2 is energetically less favored (ΔG‡ = 16.2 kcal/mol, see Supporting Information File 1 for details of the explored chlorination and bromination mechanisms). Additionally, for this mechanism, we identified that the formation of TS4 has the highest energy barrier (ΔG‡ = 20.2 kcal/mol), becoming a less probable route. This result also confirms the relevance of using two equivalents of aluminum chloride.

The aromatic chlorination with iodine(III) reagents broadly employs PhICl2. Thus, we explored two alternatives for the chlorination of 2-naphthol to identify or rule out this potential reaction pathway. The first explored mechanism involves PIFA/AlCl3 and the second PIDA/AlCl3 (see Figures S2 and S3, respectively, Supporting Information File 1). In both cases, the route involves the formation of PhICl2 as the chlorinating reagent by considering two equivalents of AlCl3 (PIFA/AlCl3 or PIDA/AlCl3, 1:2). Overall, we characterized four transition states along the reaction coordinates for both pathways. Although the PIFA-assisted mechanism follows a similar route to that described in Figure 1 until the formation of the active chlorinating species, in this case, the formation of TS2 requires 18.1 kcal/mol, which is an energetically more demanding process than the equilibration between I-2–Cl and ion pair I-3–Cl, proposed as active chlorinating species in Figure 1 and requiring less than 1 kcal/mol. It is worth mentioning that once PhICl2 is formed, the energy barrier to TS3 is 21.5 kcal/mol. These energy differences suggest that the traceroute PhICl2 is less viable for the chlorination of 2-naphthol.

On the other hand, in the presence of PIDA (Figure S3, Supporting Information File 1), when the reaction occurs through the chlorinating species PhICl2, we found that TS1, TS2, and TS4 require 17.7, 13.8, and 16.5 kcal/mol, respectively. Considering the high transition-state energy barrier in the proposed mechanism shown in Figure 1 for TS2–Cl (11.1 kcal/mol), this route is less probable. Additionally, we observed that chlorination of naphthol (the formation of I-6) could be the determining step since we found a coupling between the ring of the chlorinating species and naphthol during TS4, i.e., it could disfavor the PIDA-assisted chlorination traceroute via PhICl2. Thus, using PIFA and two equivalents of AlCl3 resulted in the highest yield, which is in agreement with our experiments.

As a consequence of the previous analysis, the chlorination process is energetically favored in the presence of PIFA/AlCl3, 1:2 through the formation of PhICl–TFAO–AlCl3 in equilibrium with [PhICl][TFAO–AlCl3] as chlorinating species.

Bromination mechanism

The reaction mechanism for the bromination of 2-naphthol using one equivalent of PIDA and two equivalents of aluminum bromide is shown in Scheme 5.

Scheme 5: Calculated reaction mechanism for the bromination of 2-naphthol using the PIDA/AlBr3, 1:2 system.

PIDA coordinates the first equivalent of aluminum bromide to form the adduct PIDA–AlBr3, which spontaneously dissociates, giving ion pair I-1–Br. Next, via TS1–Br, complex AcO–AlBr2 is released with subsequent formation of intermediate I-2–Br. Then, the second equivalent of aluminum bromide coordinates to an acetate ligand, forming adduct I-3–Br, which is the brominating species. At this point, 2-naphthol reacts and forms I-4–Br via TS2–Br. Afterwards, the second equivalent of the AcO–AlBr2 complex and HBr are released with concomitant formation of I-5–Br trough TS2–Br. Then, I-5–Br spontaneously isomerizes to give I-6–Br via the transition state TS4–Br. Finally, I-6–Br tautomerizes, yielding the experimentally observed 1-bromo-2-naphthol (P–Br).

Based on our calculations, the bromination reaction proceeds through a stepwise mechanism. Thus, the reaction starts with the coordination of aluminum bromide to an acetate ligand in PIDA to form the PIDA–AlBr3 adduct in a highly exergonic process. Similar to the previous section, the Gibbs free energy at this point was set as 0 kcal/mol for reference. At this stage, the PIDA–AlBr3 adduct undergoes ionization, giving rise to the corresponding ion pair I-1–Br (ΔG = −31.3 kcal/mol) in a highly exergonic and energetically favorable process. Next, an intramolecular SN2 reaction of the formed aluminum anion transfers a bromine atom to the electrophilic iodine(III) center through TS1–Br, which has a feasible energy barrier of 8.3 kcal/mol. The I–Br and Br–Al bond lengths are 3.15 and 2.78 Å, respectively, and the I–Br–Al angle is 93.1o, which is close to the common T-shape of such hypervalent iodine(III) species. This step releases the tetracoordinate AcO–AlBr2 salt and gives rise to the intermediate I-2–Br (ΔG = −9 kcal/mol), which contains the key Br–I(III) bond with a length of 2.65 Å. Herein, we could identify an energetically favored AcO/Br ligand exchange that releases 35.2 kcal/mol. At this point, the second equivalent of aluminum bromide is coordinated by an acetate ligand to produce the active brominating species Br–I(Ph)–OAc–AlBr3 (I-3–Br). Then, 2-naphthol adds to the iodine(III) species to release the activated Br3Al–OAc ligand through transition state TS2–Br (ΔG‡ = 11.7 kcal/mol), which leads to the protonated intermediate I-4–Br. The next step is a deprotonation assisted by the released Br3Al–OAc species. This allows the formation of the AcO–AlBr2 salt via TS3–Br and the trans intermediate I-5–Br, which contains a Br–I(Ph)–O–naphthyl bond of 2.14 Å length. The last step is the bromination of I-5–Br by isomerization to the cis transition state TS4–Br (ΔG‡ = 16.1 kcal/mol), which yields the brominated nonaromatic intermediate I-6–Br in a highly exothermic step (ΔG = −52.3 kcal/mol). Finally, I-6–Br undergoes spontaneous aromatization, converting it into the experimentally observed 1-bromo-2-naphthol (P–Br), which is more stable than I-6–Br by 2.6 kcal/mol (Figure 2).

Figure 2: Calculated mechanism for the bromination of 2-naphthol in the presence of PIDA and AlBr3.

We also explored the plausible mechanisms for the bromination of 2-naphthol mediated by PIFA in a 1:2 ratio (Figure S4, Supporting Information File 1). In this proposal, we observed that the energy barriers to reach TS1 (10.6 kcal/mol) and TS2 (16.7 kcal/mol) are higher than those calculated for the mechanism shown in Figure 2, namely 8.3 kcal/mol for TS1–Br and 16.1 kcal/mol for TS4–Br.

Other possible mechanisms involve the formation of PhIBr2. For these scenarios, the reaction pathway with PIDA and PIFA, respectively, involves two equivalents of AlBr3 (Figures S5 and S6, Supporting Information File 1). Calculations indicated that each of these pathways proceeds along four transition states. Moreover, we found that the coordination of AlBr3 to I-2 to form TS2 has the highest energy barrier (determining step, ΔG‡ for TS2 53.3 kcal/mol) in the presence of PIDA. Meanwhile, formation of TS3 (ΔG‡ = 21.3 kcal/mol) is the limiting step of the mechanism in the presence of PIFA.

Then, in the presence of the PIFA/AlBr3 system, bromination of 2-naphthol is the energetically most favored pathway. Although all these reactions occur through four transition states, significant energy differences exist concerning the PIDA/AlBr3 system. For example, the activation barrier of TS2 was 41.6 kcal/mol higher in energy than that in the mechanism in Figure 2. A similar energy profile was obtained for the bromination of 2-naphthol in the presence of PIFA (and 2 equivalents of AlBr3) compared to Figure 2. The energy difference from I-1 to TS1 (2.3 kcal/mol for the reaction in the presence of PIFA) could be the reason why the experimental yield is higher in the presence of PIDA and two equivalents AlBr3 rather than PIFA and two equivalents AlBr3 when considering the different hypervalent iodine reagents for this reaction.

To find the correlations between experiments and theoretical calculations, chlorination and bromination of 2-naphthol using PIFA/AlCl3 and PIDA/AlBr3 were carried out. Consequently, we found an excellent correlation between the yield and the energy barrier (Table 1).

Table 1: Experimental yield by using different hypervalent iodine(III) reagents.

reaction PIFA PIDA AlX3 chlorination 63% 48% AlCl3 bromination 84% 93% AlBr3